JP4736677B2 - Obstacle detection device - Google Patents

Description

  The present invention relates to an obstacle detection apparatus that detects an obstacle behind and behind a vehicle using at least a radio wave radar.

  A system that detects an obstacle around a vehicle using a radio wave radar or an image sensor is known (for example, see Patent Document 1). Based on the detection result, by performing notification to the driver and vehicle behavior control, automatic driving is performed, collision avoidance to an obstacle is promoted, and safety is ensured.

The technology described in Patent Document 1 relates to a peripheral monitoring device including a millimeter wave radar and an image sensor, and detects peripheral objects by combining data that is good according to both detection ranges and characteristics thereof. Is what you do. This technique detects, for example, a preceding vehicle when performing automatic traveling following the preceding vehicle, and the millimeter wave radar is responsible for measuring the preceding vehicle at a relatively long distance.
JP 2001-99930 A

  By the way, as such an object detection device around the vehicle, there is a system in which not only the front but also the side and the rear are used as detection areas. In a system in which a vehicle front is used as a detection area, it is used for control such as follow-up running and lane departure monitoring, and is often controlled when moving forward, particularly during high-speed running. In addition, the target object is usually a target object (preceding vehicle, oncoming vehicle) to be visually recognized within the driver's visual recognition area, and assists / reinforces the driver's visual recognition.

  On the other hand, in particular, in a detection device that uses the rear and rear sides as the target area, it is necessary to perform control not only when moving forward, but also when moving backward and when stopping. The target object also covers an area where the driver cannot visually recognize or an area where visual recognition is difficult. Therefore, it is necessary to perform a configuration and control different from the monitoring device for the front area.

  Then, this invention makes it a subject to provide the obstruction detection apparatus which has a structure suitable for the obstruction detection of the back and rear side of a vehicle.

An obstacle detection apparatus comprising a radio radar having detection areas in the rear and rear sides of a vehicle and a determination means for determining the presence / absence of an obstacle based on the detection result of the radio radar, and detecting a traveling direction in the longitudinal direction of the vehicle And a traveling direction detecting means for determining that the detected object is an obstacle when the reception intensity of the radio wave radar is greater than a predetermined threshold and is continuously detected for a predetermined tracking time or longer. When the traveling direction detecting means detects that the vehicle is not moving backward, the first threshold is used as the threshold, and when it is detected that the vehicle is moving backward, the first threshold is used. When a smaller second threshold is used as the threshold and the second threshold is used as the threshold, the tracking time is set longer than when the first threshold is used. It is characterized in.

  When the vehicle is not moving backward (including not only moving forward but also stopping), only objects moving toward the vehicle may contact or collide with the vehicle from the rear or rear side. Usually, it is limited to the following vehicle. On the other hand, during reverse, a stationary object on the track, an object moving across the track (including other vehicles and pedestrians), or an object that is slower than the host vehicle even if moving in the track direction It can become an obstacle that may contact and collide. Radio radars are good at detecting metallic objects such as vehicles, but are relatively weak at detecting non-metallic objects such as stationary objects and low-speed objects. Therefore, in the present invention, this kind of object can be easily detected by loosening the obstacle determination condition when retreating.

By setting the threshold value for determining an obstacle as low when the vehicle is moving backward, it is easy to determine the obstacle when moving backward.

If the tracking time is lengthened, it takes time to determine the obstacle. However, there is a high possibility that the detection target at the time of backward movement stays in the detection area for a long time as compared with the time of forward movement, and the detection is ensured by extending the tracking time.

  While the vehicle is moving forward and stopped, the relative approach speed is positive.In other words, there is little possibility of approach / collision except for an object approaching the host vehicle, so there is no need to determine that it is an obstacle. When the vehicle is not moving backward, only the detected object having a positive relative approach speed may be determined as an obstacle. Whether it is an approaching object or not can be easily separated using the Doppler component of the reflected wave.

  An image sensor that overlaps at least a part of the detection area with the radio radar is further provided, and the determination means detects an object in the overlapped detection area with the radio radar when the vehicle is moving backward. The object may be determined as an obstacle when the object is also detected by the image sensor, and the object detected by the radio wave radar may be determined as an obstacle when the vehicle is not moving backward.

  As described above, radio wave radar is good at detecting metal objects such as vehicles, but is relatively weak at detecting non-metal objects. Therefore, at the time of retreat, the detection condition of the obstacle in the radio wave radar is relaxed, so that detection of a non-metallic object is possible, and erroneous detection is suppressed by combining the detection results of the image sensor. On the other hand, at the time of advancement and stop, in order to give priority to the detection of the metal object, detection by the radio wave radar is performed using a judgment condition that is not relaxed.

  According to the present invention, because the obstacle is detected under a gentler judgment condition than when the vehicle is moving backward while stopping or moving forward, an object that the radio wave radar is not good at detecting, for example, a pedestrian or a non-metallic stop It becomes possible to determine the presence or absence of an object or the like. For example, if this is dealt with by changing the determination threshold value, it is not necessary to greatly change the determination logic, and it is easy to realize.

  It is desirable to loosen the determination threshold value when retreating, and to lengthen the tracking time, since it is possible to eliminate a temporary influence such as reflection due to road surface unevenness and improve the determination accuracy. Even in this case, since the moving speed at the time of backward movement is relatively low, it is possible to secure a time margin until contact / collision.

  When the vehicle is not moving backward and is moving forward or stopped, only obstacles approaching the vehicle need be determined. When processing such as vehicle behavior control is performed according to an obstacle, it is not necessary to transmit an object with low possibility of contact as information to such control processing, so that the memory and calculation amount required for the processing can be reduced. .

  When image sensors that partially overlap the detection area are provided, only the object that can be detected by both the radio radar and the image sensor is used as an obstacle for the overlap area only when the vehicle is moving backward. This makes it possible to accurately detect an object that is particularly difficult to detect with a radio wave radar.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the accompanying drawings. In order to facilitate the understanding of the description, the same reference numerals are given to the same components in the drawings as much as possible, and duplicate descriptions are omitted.

  FIG. 1 is a block diagram showing a configuration of an embodiment of an obstacle detection apparatus according to the present invention. The obstacle detection apparatus 100 includes an obstacle recognition ECU 1, a rear camera 2 that acquires an image behind the vehicle, a radar 30 that searches backward from the right rear of the vehicle, and a radar 31 that searches backward from the left rear. . The radar 31 is preferably a radio wave radar such as a millimeter wave radar or a microwave radar. Radio radar recognizes objects even in environments where it is difficult to detect with the rear camera 2 because it can perform well without being affected by weather conditions such as rain, fog, snow, ambient brightness, dust, and noise. can do.

  The obstacle recognition ECU 1 includes a CPU, a ROM, a RAM, and the like, and includes an image processing unit 11 that performs image processing of an image acquired by the rear camera 2 and a radar processing unit 12 that processes search results of the radars 30 and 31. And a determination unit 14 that determines the presence or absence of an obstacle based on the processing result, and a storage unit 13 that accumulates information necessary for the determination. The image processing unit 11, the radar processing unit 12, the storage unit 13, and the determination unit 14 in the obstacle recognition ECU 1 are not limited to completely independent hardware, and part or all of them are other processing units and units. And some or all of the hardware may be shared. That is, the same hardware can be realized by a plurality or one software.

  In addition to the processing result of the radar processing unit 12 being input to the image processing unit 11, the determination unit 14 is input with the determination result of the image processing unit 11 and the radar processing unit 12 and is stored from the storage unit 13. The output of the shift sensor 70 for detecting the shift state of the drive system and the output of the vehicle speed sensor 71 for detecting the vehicle speed are input.

  The obstacle recognition ECU 1 is connected to the monitor 51, the speaker 52, the vehicle behavior control ECU 53, the engine ECU 54, the pre-crash system ECU 55, and the like via the in-vehicle LAN 4. Each of the ECUs 53 to 55 is constituted by a CPU, a ROM, a RAM and the like in the same manner as the obstacle recognition ECU 1, and controls the behavior of the vehicle, the driving of the engine, and the operation of the pre-crash system, respectively.

  FIG. 2 is a diagram illustrating detection areas of the rear camera 2 and the radars 30 and 31. Each of the radars 30 and 31 is disposed at the left and right corners behind the vehicle 6, and uses a fan-shaped area having a predetermined angle as a detection area. On the other hand, the rear camera 2 is installed at a substantially central portion at the rear end of the vehicle 6 and captures an image within a predetermined angle of view, and the imaging range is a detection area.

  Here, a detection area that overlaps all of the rear camera 2 and the radars 30 and 31 is defined as an area a, and an area that can be detected only by the rear camera 2 can be detected by overlapping the area b and the rear camera 2 and the radar 30. The area outside the detection area of the radar 31 is area c, and the rear camera 2 and the radar 31 can be detected redundantly and the area outside the detection area of the radar 30 is the area d, and the radar 30 and the radar 31 are The area that can be detected redundantly and is out of the detection area of the rear camera is area e, the area that can be detected only by the radar 30 is area f, the area that can be detected only by the radar 31 is area g, and the rear camera 2 and the radar 30 , 31, an area located on the right front side of the area f and outside the detection area 31 is called an area h, and an area located on the left front side of the area g is called an area i. .

  Here, some forms of the operation of the present embodiment will be specifically described. FIG. 3 is a flowchart showing the first control process. In the first control process, an obstacle behind the vehicle is determined based on the detection results of the radars 30 and 31. This processing is repeatedly executed at a predetermined timing by the determination unit 14 in the obstacle recognition ECU 1 while the vehicle is powered on.

  In step S1, the detection result of the radar processing unit 12 is read. Subsequently, the vehicle traveling direction detection result is determined from the output signals of the vehicle speed sensor 71 and the shift sensor 70 (step S3). In step S5, it is determined whether the vehicle is moving backward. When the vehicle is moving backward (when it is determined that the vehicle is actually moving backward at a predetermined speed or more), the process proceeds to step S7, where the intensity MP of the reflected wave detected by the radar exceeds a predetermined threshold value MPth1. It is determined whether or not. As this MPth1, a threshold value is set which can distinguish a reflected wave from a metal object such as a vehicle from other reflected waves and noise.

  When MP is larger than MPth1, the process proceeds to step S9 and 1 is added to the counter value tc. The initial value of tc is set to 0, and since there are a plurality of obstacles, when there are a large number of reflected waves, they are counted individually for each corresponding reflected wave. When the obstacle is continuously detected, the continuous time (tracking time) is counted, and the actual tracking time can be expressed by time between time steps × tc.

  In step S11, tc is compared with a threshold value tc_th1. If tc exceeds tc_th1, the object is determined to be an obstacle (step S13), and the process ends. If tc is equal to or less than tc_th1, the processing is terminated as it is. In this case, the determination as to whether or not the object is an obstacle is suspended.

  On the other hand, if it is determined in step S7 that MP is equal to or less than MPth1, the process proceeds to step S15, and MP is compared with another threshold value MPth2. Here, MPth2 is set to a value smaller than MPth1, and for example, a reflected wave corresponding to a reflected wave from a pedestrian, a non-metallic obstacle, concrete, wood, or the like is set to a value that can be identified from noise or the like. ing. When MP is larger than MPth2, the process proceeds to step S17 and 1 is added to the counter value tc. Then, this tc is compared with a threshold value tc_th2. This tc_th2 is set sufficiently longer than the above-described tc_th1. If tc exceeds tc_th2, the object is determined to be an obstacle (step S13), and the process ends. By reducing the threshold value of reflected wave intensity, it is possible to detect non-metallic objects with weak reflected wave intensity, but there is also the possibility of misrecognition that the obstacle is judged even though there is no obstacle. Increase. In this control process, while the threshold value of the reflected wave intensity is lowered, the occurrence of this erroneous recognition is suppressed by setting the tracking time threshold value long. If tc is equal to or less than tc_th1, the processing is terminated as it is. In this case, the determination as to whether or not the object is an obstacle is suspended.

  If it is determined in step S15 that MP is equal to or less than MPth2, the process proceeds to step S23, the counter value tc is reset to 0, it is determined that there is no obstacle (step S25), and the process ends.

  If it is determined in step S5 that the vehicle is not moving backward (that is, moving forward or stopped), the process proceeds to step S27, and only the detection result of the object approaching the host vehicle is extracted (specifically An example of the process will be described later). This means adding to the obstacle determination condition that the vehicle is approaching when not moving backward, and the condition for determining an obstacle is stricter than when moving backward, in other words, This means that it is easier to determine an obstacle when the vehicle is moving backward than in other cases. Then, the process proceeds to step S29, where it is determined whether or not the extracted reflected wave intensity MP exceeds a predetermined threshold value MPth1.

  If MP is greater than MPth1, the counter value tc is incremented by 1 (step S31). Then, tc is compared with a threshold value tc_th1 (step S33). If tc exceeds tc_th1, the object is determined to be an obstacle (step S35), and the process ends. If tc is equal to or less than tc_th1, the processing is terminated as it is. In this case, the determination as to whether or not the object is an obstacle is suspended.

  If it is determined in step S29 that MP is equal to or less than MPth1, the process proceeds to step S37, the counter value tc is reset to 0, it is determined that there is no obstacle (step S39), and the process ends.

  In this way, by extracting only the approaching object while moving forward or stopping and performing obstacle determination, discrimination processing for a stopped object that does not hinder the traveling of the host vehicle or an object that moves away from the host vehicle is performed. This can be omitted and the amount of processing can be reduced. Thereby, the process according to an obstacle, for example, a vehicle behavior can be controlled according to the situation of a parallel running vehicle, a following vehicle, etc., or the process at the time of operating a pre-crash system can be performed rapidly.

  Next, the second control process will be described with reference to the flowchart shown in FIG. The second control processing is different from the first processing mode in that the processing is performed using both the detection results of the radars 30 and 31 and the processing result obtained by performing image processing on the image acquired by the rear camera 2.

  Here, together with the reading of the radar detection result (step S1) and the reading of the traveling direction detection result (step S3), the image processing unit 11 reads the image processing result (step S2). These processing results may be read in parallel or in a different order.

  The reverse determination processing in step S5 and the processing when it is determined that the vehicle is not in reverse (steps S29 to S39) are the same as those in the first processing mode. If it is determined in step S5 that the vehicle is moving backward, a comparison process between the intensity MP of the reflected wave detected by the radar and the threshold value MPth1 is performed as in the first processing mode (step S7). Here, the process when MP exceeds MPth1 (steps S9 to S13) is the same as the first process form. On the other hand, if MP is equal to or less than MPth1, MP is compared with another threshold value MPth2 as in the first processing mode (step S15).

  The processing (steps S23 and S25) performed when it is determined in step S15 that MP is equal to or less than MPth2 is the same as the first processing mode. On the other hand, if it is determined in step S15 that MP exceeds MPth2, the process proceeds to step S16 to determine whether the image detection result and the radar detection result match. Specifically, it may be determined whether there is an obstacle candidate detected by image processing at a position corresponding to the reflected wave. If it is determined that they match, the process proceeds to step S21, where it is determined as an obstacle, and the process ends. If they do not match, the process ends. In this case, the determination as to whether the object is an obstacle is suspended.

  In this processing mode, when the vehicle moves backward, the threshold value of the reflected wave intensity of the radar is made lower than in other cases so that a non-metallic object or the like can be detected. Since the possibility of misrecognition increases by lowering the threshold value, in order to suppress this misrecognition, the processing result of image recognition is used together for an object with low reflection intensity. The vehicle speed when reversing is relatively low, and objects with low reflection intensity include pedestrians, non-metallic walls, trees, etc., but these are relatively large in size and are they stopped? Since the speed is slow, the rear camera 2 can continuously capture the image, and if it can be an obstacle that affects the progress of the vehicle, the size on the screen is relatively large. Therefore, it is relatively easy to grasp by image recognition.

  Here, the approaching object extraction process in step S27 will be described. As an approaching object extraction method, for example, by detecting the Doppler shift component of the radars 30 and 31, the relative speed between the object and the host vehicle can be known. When this is applied to the first processing mode, an obstacle can be detected based only on the processing results of the radars 30 and 31.

  Further, the approaching object can be detected by combining the processing results of the radars 30 and 31 and the image processing result. FIG. 5 is a flowchart showing the extraction process in that case. In step S51, it is determined whether the radar detection object is within the image detection area. If the detected object is outside the image detection area, the process ends. In this case, no extraction is performed. When the radar detection object exists in the image detection area, the process proceeds to step S53, and the image processing unit 11 performs image processing on the periphery of the radar detection object to determine the size and shape of the object. In step S55, the image processing result and the radar detection result are merged, and it is determined whether or not the object is approaching the host vehicle from the change in position, size, shape, and the like.

  Next, the third control process will be described with reference to the flowchart shown in FIG. The second control processing is different from the first processing mode in that the processing is performed using both the detection results of the radars 30 and 31 and the processing result obtained by performing image processing on the image acquired by the rear camera 2.

  First, the presence / absence of a radar detected object is determined (step S61). When detecting the presence / absence of a radar detection object, as described in the first and second control processes, the judgment threshold value is changed between backward and other cases, and the object is detected during backward movement. It is good to set it easily.

  If there is a detected object, the process proceeds to step S63, and it is determined whether or not the detection position is within the image detection area. If it is in the image detection area (in the case of areas a, c, and d in FIG. 2), the process proceeds to step S65 to determine whether or not the object is an obstacle by performing image processing around the radar detection object. To do. If it is not within the image detection area (if it is within the area e, area f, or area g in FIG. 2), the object detected by the radar is set as an obstacle (step S67). At this time, the tracking time threshold value may be changed according to the intensity of the reflected wave as in the first control process described above.

  If it is determined in step S61 that there is no object detected by the radar, the process proceeds to step S69 to determine whether there is an obstacle detected last time. If no obstacle has been detected in the previous time, it is determined that there is no obstacle in the vicinity, and the process is terminated.

  On the other hand, if an obstacle has been detected last time, the process proceeds to step S71 to determine whether or not the predicted position is within the image detection area. Here, the position prediction may be performed by tracking the position of the obstacle and predicting the current relative position based on the previous relative position, relative speed, and relative acceleration information of the obstacle. The case where the predicted position is located in the image detection region is a case where the predicted position is considered to be located in the area b in FIG. This is because the obstacle is not detected by the radar, and is not considered to be located in the areas a, c, and d that can be detected by the radar. When it is determined that the object is located within the image detection area, an obstacle is determined by performing image processing around the predicted position (step S73). The case where the predicted position is not within the image detection area is a case where the predicted position is considered to be located in an area where neither the radar nor the image can be detected, that is, the area h or the area i. In this case, it is estimated that there is an obstacle at the predicted position, and the estimation result is output as obstacle information (step S75).

  In particular, when the vehicle is moving forward, it recognizes the traveling lane and includes an approaching obstacle in the traveling lane (including a vehicle entering from another lane in addition to the following vehicle traveling on the traveling lane). ) May be recognized as an obstacle.

It is a block diagram which shows the structure of embodiment of the obstruction detection apparatus which concerns on this invention. It is a figure which shows the detection area of each of the rear camera 2 and radar 30,31. It is a flowchart which shows the 1st control processing of the apparatus of FIG. It is a flowchart which shows the 2nd control processing of the apparatus of FIG. It is a flowchart which shows an example of the determination process of an approaching object. It is a flowchart which shows the 3rd control processing of the apparatus of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Obstacle recognition ECU, 2 ... Back camera, 4 ... In-vehicle LAN, 6 ... Vehicle, 11 ... Image processing part, 12 ... Radar processing part, 13 ... Memory | storage means, 14 ... Determination part, 30, 31 ... Radar, 51 DESCRIPTION OF SYMBOLS ... Monitor, 52 ... Speaker, 53 ... Vehicle behavior control ECU, 54 ... Engine ECU, 55 ... Pre-crash system ECU, 70 ... Shift sensor, 71 ... Vehicle speed sensor, 100 ... Obstacle detection apparatus.

Claims (3)

In the obstacle detection device comprising a radio radar having a detection area in the rear and rear sides of the vehicle, and a determination means for determining the presence or absence of an obstacle based on the detection result of the radio radar,
A traveling direction detecting means for detecting a traveling direction in the longitudinal direction of the vehicle;
The determination means determines that the detected object is an obstacle when the reception intensity of the radio wave radar is larger than a predetermined threshold and is continuously detected for a predetermined tracking time or longer, When the direction detecting means detects that the vehicle is not moving backward, the first threshold value is used as the threshold value, and when it is detected that the vehicle is moving backward, the second threshold value is smaller than the first threshold value. Is used as the threshold value, and when the second threshold value is used as the threshold value, the obstacle detection is characterized in that the tracking time is set longer than when the first threshold value is used. apparatus.   2. The obstacle detection apparatus according to claim 1, wherein the determination means determines that an object having a positive relative approach speed is an obstacle at least when the vehicle is not moving backward. It further comprises an image sensor that overlaps at least a part of the detection area with the radio wave radar,
In the case where the object in the overlapping detection area is moving backward, the determination means determines that the object is an obstacle when the object detected by the radio wave radar is also detected by the image sensor. judgment, when the vehicle is not in retreat, the obstacle detection apparatus according to claim 1 or 2, characterized in that to determine the target object detected by the radio wave radar and an obstacle.

Images